Month: September 2021

Categories
Products and Design The Next Computer

Everything is on high-speed internet. Why are we seeing wait spinners all the time?

In this blog post about offline-first, this important point:

Latency is more important than bandwidth

In the past, often the bandwidth was the limiting factor on determining the loading time of an application. But while bandwidth has improved over the years, latency became the limiting factor. You can always increase the bandwidth by setting up more cables or sending more Starlink satelites to space. But reducing the latency is not so easy…

Offline first applications benefit from that because sending the inital state to the client can be done much faster with more bandwidth. And once the data is there, we do no longer have to care about the latency to the backend server.

This is why iOS Background App Refresh, when implemented well, works like magic. iCloud Tabs in Safari are a great example of this:

Safari doesn’t stop you from using the browser while it syncs tabs. In fact, if it detects that the connection isn’t good enough to fetch tabs from other devices quickly enough, it just won’t show you the “From {device name}” section.

Safari continues to attempt to sync tabs in the background, where you’re using the browser or not. When it’s synced, the section shows up. Like magic.

iCloud photos is another example of offline first; there is never any wait time while you use the same photo library across multiple iOS and Mac devices.

Likewise the podcast app Overcast will download podcasts and sync subscriptions silently in the background.

In both cases, the developers have designed for the fact that bandwidth over time is abundant, but when the user launches the app, (lack of) latency is important, so the apps don’t sync right then.

Of course, conflict resolution is an important part of offline-first, and today many applications do this at a file level. The interface’ll ask you which version (from another device or from the local device) you want to use.

Other, more intelligent applications will do this within a file. A text editor like TextMate on MacOS detects that a file that is open in the editor has been changed on the filesystem because it was edited on another device and has synced via, say, iCloud or Dropbox. The editor then uses markup to highlight the conflicting text.

Either way, the application won’t stop you from using it until it can determine everything is synced, which often isn’t possible quickly.

Unfortunately, too many apps today rely on calling home at every launch, even if it isn’t to sync files. Compare the launch of the image creation app Canva and the Twitter app Tweetbot (both cold launches). You can see which one loads existing content first and then start looking to sync. Its clear which one feels more snappy:

Canva
Tweetbot

Whether your users are on a phone perpetually connected to 4G, a desktop plugged into gigabit Ethernet, a laptop with patchy wifi, or a tablet with no available connections nearby, an offline first app gives them confidence that it’ll be available instantly when they need it. To capture text. Review a photo. Add a contact. Or even browse the web.

(ends)

Categories
The Next Computer

iOS apps worth trying out

Recently someone on Twitter bought the new iPhone 13 after several years on Android, and asked about iPhone apps worth trying. This is what I sent them. Here’s hoping you find this list useful.

Read/Social

Productivity 1

Productivity 2

Photos/Videos

Other utilities

Categories
Life Design

The disrupted and the deliberate

I used to think that disruptions in habits were temporary. One things got back to normal, I’d automatically go back to my regular schedule. I’ve discovered quite powerfully that that isn’t necessarily the case.

I fell ill in April 2018 for a couple of weeks and never really got back to my workout routine that I had built up over years, a routine I persisted with even through mental illness.

Earlier this year, some major disruptions caused changes to my sleep schedule. The disruption has passed but my sleep schedule has remained shifted by a few hours. This was a schedule I’d developed as a child and stuck through adulthood to middle age.

And also in 2018, I finished no books, an extraordinary feat for me – I’d read at least a couple of dozen books a year since my teens. But I’ve found it hard since to get absorbed in a book from start to finish – and I had, until now, known no other way.

Just like it had been with sleeping early, quickly and well, just like with being in shape had.

I’m reconstructing each of those habits once again. There are several frameworks to help, such as Atomic Habits. More than anything else, though, is the hard realisation that each of these is going to be a journey, one that requires me to be deliberate at and with each step. Habits that are knocked out of orbit by disruptions don’t just wobble back into their old trajectories, they need to be trained to revolve around our selves once again.

Categories
Product Management Startups

Desire-friction mismatch

Today

This reminded me of something: for ten years now, I have dealt with or advised founders about desire-friction mismatch.

Founders often assume that because they have built a particular product that a certain defined market segment desires – ie because they have achieved product market fit, that their numbers should now take off.

It doesn’t always.

PMF is an important milestone. But achieving it doesn’t tell a founder how motivated the average person in their market segment is. It doesn’t tell them how their promise tips the stakes for average would-be customers.

At a certain level of friction, customers’ desire to start using your product won’t be enough. They’ll sigh and move on.

At this point, a founder needs to engage their product person. To take whatever minimum viable product that got them to product-market-fit and iron the wrinkles out of the customers experience. These wrinkles exist as a result of compromises made – rightly – to speed up the initial go-to-market.

But now the founder needs to invest in the product so that it’s simple to understand and easy to get started with for the larger-than-earlier numbers of customers to come.

Unfortunately, it’s at this very point that a founder’s attention and investment turns from product (and tech) to sales and marketing in an attempt to acquire customers at a larger scale than during the PMF phase. To now look for sustainable, efficient acquisition/distribution channels.

What can happen – and too often does – is that a startup spends more (money, time, attention, emotion) in acquiring lots of new leads and sending them down a signup flow where the friction outweighs their desire to get through it.

When numbers don’t ramp up as quickly as expected, the founder is now unsure about the product-market-fit itself. They think about what they got wrong. Maybe they revisit it. Maybe they have the product people make cosmetic changes to the product trying to communicate the value proposition even more loudly (too often by adding more text/visual clutter).

In this case the problem isn’t that customers don’t get the value proposition. They do – the team’s now past the PMF milestone. But they don’t have the motivation to deal with the friction that the still-early product presents – not every product is as important to people’s lives as play-to-earn games like Axie are to people in the Philippines.

Being able to diagnose if it’s lack of PMF or desire-friction mismatch is often what makes the difference between an excellent startup’s post-PMF journey taking off, and sputtering.

Categories
The Next Computer

What we think of as a computer has changed in ways our 2007 selves could not imagine

Several of the reviews of this year’s iPhone 13 line from mainstream USA publications were lukewarm about the main decision a reader wanted to make: should they buy the new iPhone or not?

Source: this Youtube comparison video

CNN was clear you should only upgrade if you’re on an old iPhone (or are a professional photographer). The New York Times began with the sentence “the truth is that smartphones peaked a few years ago”. CNET compares the newest phones to older ones and only starts suggesting upgrades with phones three years old.

It’s clear that people are waiting longer to get a new smartphone than they did several years ago. And that wait is getting longer.

I think it’s mostly because smartphones – now just phones – have become an everyday essential. Far from the novelty they were for their first decade, they are now how we interact with the world. Consequently, the choice to get a new iPhone is no longer about signalling exclusivity or wealth; it’s a much more utilitarian decision: whether the phone one owns is now long in the tooth.

And the answer to that question is usually No. For a few years now, phones have been much more capable than the average person needs them to be. Give the average person an iPhone XS instead of the iPhone 13 from three generations later, and they’ll have a hard time convincing themselves that the new one is really all that better.

Phone processors are also as fast as most desktops and laptops. Graphics performance is arguably better. They also have about the same amount of storage space. We use them for different things than we do our computers, but they’re really – as Apple describes its iPad – the Next Computer. It’s just that in ten years, what we think of as a computer has changed in ways our 2007 selves could not imagine.

So it stands to reason that phones would cost as much. In the USA, iPhones used to cost around $500. The first iPhone cost $499 and $599 depending on storage. The iPhone 13 Pro Max, the biggest model in the fifteenth generation of iPhones, now starts at $1099. Only a few Android phones are as expensive, but the percentage increase in price is about the same.

That’s as much as a laptop computer. And so therefore if phones are already faster than we need them to be, they play the same role in our lives that our laptops used to, and they cost as much as a laptop, we shouldn’t be surprised that we replace them as (in)frequently as we do our computers – that is, once every three or four years.

I’d expect that gap to grow.

Apple has understood this for a long time now.

Years ago, they publicly said they were ok with the iPhone cannibalising iPod revenue (because they’d rather they themselves did than anyone else). In much the same way, they welcome, even encourage the use of iPhones for years on end. And in balance, they have opened up opportunities to earn throughout that period of ownership, through services: iCloud. Apple TV+. Apple Music. Arcade. Fitness+. And very likely more in the future.

from Apple

The best way to predict the future is to invent it.

Categories
Uncategorized

And yet it seems smaller

In the nineties, the internet seemed to be a vast endless expanse of wonder and weirdness. Because it was. Especially before any localisation or personalisation.

In the nearly thirty years since, the internet, particularly the web, has exploded, and now it’s orders of magnitude bigger: more people from more countries, more domains, more types of stuff online, more types of devices – and way faster pipes.

And yet it seems smaller. The relentless optimisation by search engines and social media (well, Google and Facebook) has de-weirded the Internet, stripped it of its long tail. Everyone’s talking, but it’s about the same set of things.

I think we need more entry points to the Internet to get some of that magic back. Social media doesn’t cut it, not even Reddit, at one time the Front Page of the Internet. And metafilter survives valiantly, but as a shadowy of its former self (or perhaps in the shadow of what is now the giant corporate internet).

What are some worthy successors to StumbleUpon, Digg and del.icio.us?

Categories
Uncategorized

What value does the token capture?

A friend brought to my notice this important comment, ostensibly about crypto tokens:

https://twitter.com/benedictevans/status/1430149538006851584

same thread:

https://twitter.com/benedictevans/status/1430162049422790674?s=20

With Bitcoin and other digital-money-type cryptocurrencies, this point of view is understandable. The total mined Bitcoins are now worth nearly two trillion dollars (no one yet denominates items in Bitcoin; the US dollar has that locked down). Millions of people trade bitcoin 24×7, institutions hold it, firms create derivative instruments for it, publications cover it, social media chats constantly about it.

And yet, there are very few applications of Bitcoin outside of it being a speculative asset. It has not displaced the dollar or even threatened to do so.

Events in Syria, Afghanistan, Venezuela, Lebanon and other countries have severely devalued people’s savings in their local currency, but we have yet to see a wholesale shift to bitcoin. This is for several reasons, not necessarily a lack of trust in an all-digital currency, But the fact remains.

The decentralised finance world is similar but slightly less straightforward. Hundreds of projects doing very interesting things have their own native token. Several people do their research, identify which projects actually have teams, are building products, have pedigreed backers, and then buy and hold the native tokens of such projects. The expectation – and basis for speculation – is should the project do well, the token will rise in value too.

That isn’t necessarily true. The research should also look at what the native token represents. How much of the value that the project creates does the token capture?

It is a lubricant in transactions in the project ecosystem? In that case, how does holding the token cause its value to appreciate (other than you holding so much that you cause scarcity).

Does the token aid proof of stake? In that case, how distributed is the network? Does your holding the token aid one of just a few concentrated pools?

Does the token confer voting power on decisions? In that case, who (or what) decides what is brought to a vote or not? Are decisions binding? How diverse is the voting pool?

Did the project’s backers get some exposure to their project’s native tokens themselves? If not, why are they not aligning themselves with the project’s value capture?

This research is not radically different from that in the world of traditional finance. If you aren’t doing this, it’s fine – then do make peace with the fact that your crypto holdings and trades are just that – speculation.

Categories
Uncategorized

We’re all renting hard drives in the cloud now

Every month I get a receipt from Apple about my monthly iCloud Drive charge.

It’s a reminder of the fact that in addition to all the hard drives I’ve bought from Apple – in my computers, phone and tablet – I also rent one from them, in perpetuity.

All in order to keep data on the other hard drives in sync.

It made sense at one time. When Apple’s free iCloud tier offered 5GB of space, and they sold iPhones with 8GB of storage. Then, it was a secure, private way to store data that wouldn’t fit on your phone.

Today, ten years later, all of Apple’s iPhones and iPads start at eight times that, 64GB storage. And the latest iPhones 13, and all iPads Pro have a minimum of 128GB.

Meanwhile, the base iCloud tier is still 5GB.

Because Apple’s model is a client-server one – iCloud in the center, all of a person’s Apple devices (iPhones, iPads, Macs, Apple Watches, Apple TVs) around it – there’s no way that the base storage tier is enough for anyone’s data. And everyone ends up paying for what is effectively a rented hard drive from Apple. An intelligent one, but a hard drive nevertheless.

But most people have their devices in close proximity to each other. They’re usually connected to high speed wireless networks throughout nearly every day. An alternative model could easily have been one where each of these devices stay in sync in a mutual, peer to peer structure. If you have a don’t own a Mac, Apple could extend the Windows iCloud client to sync your data.

With this, you don’t pay for as much storage online. And Apple’s data centre scale reduces a bit. You also don’t connect to the public internet. That saves data, frees up bandwidth and is a lot more private.

A peer to peer model would fit in a lot better with Apple’s very public commitment to the environment and to privacy.

I’ve moved most of the documents I don’t use very often to a peer to peer system via Resilio Sync (for which I have a family license). I have a dozen GB synced between three Macs, one iPhone and one iPad without needing an Internet connection:

You can see that as I type this on my Mac, I have two other machines that are syncing data. All on my local LAN.
Nearly 30 GB of my iPhone’s 128 GB is my own documents, synced locally

Neither my iPhone nor iPad use iCloud Backup – they are both backed up to my Mac weekly. That saves me over fifty GB of space I would have otherwise paid for. It’ll also mean I can set up a new iPhone a lot faster than from iCloud.

Ours are the first generations to accumulate digital data through our entire lifetimes: documents, files, photos, videos, music. Each of us will easily have several terabytes of data. At some point, paying today’s rates to store all that information online is going to cost too much money.

Maybe we’ll have near-ubiquitous connectivity, essentially free bandwidth and rock-bottom rates for cloud storage, all with the guarantee of security, resilience and privacy. I’d love to see that come to pass.

Until then, I think we need to consider how we’re going to manage our data the way we did just ten years ago – ourselves, on our own machines.

Categories
Data Custody Real-World Crypto

When you own bitcoin but defeat the purpose of owning bitcoin

Paypal, Venmo and Square opened up buying and selling cryptocurrency in their apps in the last year, at least to customers in the USA.

Recently, the Canadian investing service Wealthsimple announced the ability to “buy and sell cryptocurrencies instantly”, starting with a list of about a dozen such tokens.

The benefit of such widely used services offering easy access to cryptocurrencies is clear: anyone who has been curious about bitcoin and other cryptocurrencies can now ‘buy’ them. They can experience dizzying gains and crushing losses. It’s informative and educative. And it goes beyond crypto: more people than ever can experience how this volatile an asset behaves.

At the same time, the highly abstracted nature of these experiences means that people may not understand what makes these tokens important beyond volatile curiosities. Particularly the idea of total ownership of your tokens without a mediating party – like these apps.

With Venmo, Paypal and Wealthsimple you don’t actually own bitcoin or any cryptocurrency. There’s no indication that Paypal, say, buys a certain amount of bitcoin on some exchange and holds it on your behalf. At most, it holds some bitcoin – it its name – and when you ‘buy’ bitcoin within the app for US dollars, it earmarks, in software, some of that bitcoin against your account.

No wonder none of these will let you send your bitcoin (or crypto) to another one of your own wallets. You can only sell ‘your’ bitcoin back to them (Venmo, Wealthsimple faqs).

That goes against the very basic idea of bitcoin, which was that you shouldn’t have to rely on third parties like banks to hold your money for you. There’s a high risk that ordinary customers of these apps will view the highly abstracted, processed bitcoin (and other crypto) as mere trinkets, toys. Venmo encourages you to buy bitcoin so your friends can get a notification about it:

At least Square’s Cash app gives you an option to send your holdings to an external wallet.

Beyond this, people hold cryptocurrency at the very exchanges where they buy it from, in exchange for their local currencies.

Here the situation is somewhat less bad. You do have a bitcoin wallet, or several other wallets for different cryptocurrencies: you can deposit from an outside wallet to this one. You can send tokens from this wallet to an outside one.

But your ability to operate this wallet is controlled entirely by the exchange. If it decides you have violated its terms of use, or if the exchange is attacked, or there’s a technical problem that causes it to lose your credentials, you will no longer be able to access your tokens.

Tens of millions of people – more likely hundreds of millions – hold such ‘hot’ wallets at exchanges. They all trade control for convenience.

Only a small fraction of the total number of people who have bought bitcoin or cryptocurrency tokens hold them in wallets that they truly control, in what are termed ‘non-custodial’ wallets, meaning no one holds your tokens in custody other than you yourself.

Now while you have total and exclusive control, you also bear sole responsibility for keeping your access keys and recovery passphrase safe. A significant fraction of the bitcoins that have ever been mined have been just lost because people forgot or lost what they needed to access them.

Today mediating companies like PayPal and exchanges like, say, India’s CoinDCX or the USA Coinbase, far outspend non-custodial wallets on acquiring new customers.

I think for the decentralised economy to cross any sort of realistic threshold of relevance, vastly more people need to experience holding their own tokens in non-custodial wallets. That’s when they’ll get a sense of what decentralised tokens, decentralised money means.

(ends)

Categories
Life Design Wellness when Always-On

Social connections and “a lifelong journey together”

This New York Times article from three years ago describes the degree to which strong social bonds with people who share similar interests improve one’s longevity.

One example is Okinawa, Japan, far south of the country’s main islands:

… a place where the average life expectancy for women is around 90, the oldest in the world, people form a kind of social network called a moai — a group of five friends who offer social, logistic, emotional and even financial support for a lifetime.

In a moai, the group benefits when things go well, such as by sharing a bountiful crop, and the group’s families support one another when a child gets sick or someone dies. They also appear to influence one another’s lifelong health behaviors.

“Traditionally, their parents put them into moais when they are born, and they take a lifelong journey together.”

That’s not the only reason Okinawans live longer than even the average Japanese – diet and genetics seem to play their part too – but social practices seem to be the third pillar of their longevity.

This National Geographic article from last year, which touches briefly on each of these three factors, also mentions moais.

Takashi Inafuku, head of one of Ogimi’s districts, belongs to two moai—one with a group of school friends and another with former co-workers. “They are places where you can exchange information and communicate with others,” he says. “I think that participating in moai, having a common hobby and releasing stress, can help promote longevity.”

And that

“loneliness is as bad for you as smoking.”

Image: National Geographic

While the forced isolation of the last nearly two years has had a devastating effect on people’s mental health, I have observed clearly that those with such strong social connections have fared better during this period. Often, such connections had been built and nurtured electronically even before the pandemic, and so weren’t buffeted as much by it.

(ends)